2D mm-wave imaging based on singular value decomposition

Abstract

In this paper, we develop a systematic framework for two-dimensional mm-wave imaging, based on the singular value decomposition (SVD) of the Helmholtz wave equation under the Born approximation. We identify the degrees of freedom as a function of the geometry of the aperture and the scene, and provide insight into the eigenmodes identified by the SVD. For sparse arrays with number of elements smaller than the degrees of freedom, we propose, and experimentally demonstrate the efficacy of, an eigen-filtered pseudo-inverse algorithm which selects the eigenmodes being imaged.